知识卡片
过滤样本应同步全部输入与目标
只对单独depth列过滤会改变其长度,而其他特征与目标仍保留原行。应在同一原表按统一掩码选行,再生成对齐的X与y,保留被排除索引与原因。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 使用数据
原书出处 · 1 处
- · 查看原书依据
depth=depth[depth<=max_val] depth. shape (53572,) This shows that the outliers are deleted and the total number of instances in depth is now reduced by the number of outliers. Note that the total number of instances in the Diamond data set is 53,940 and there are 366 outliers. The final number of instances reported is 53,572, which is two more. This is because there are two feature
参考来源
- 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第05章《使用数据》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/17_chapter05.xhtml)
- 结论依据:摘录中关于过滤样本应同步全部输入与目标的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:depth=depth[depth<=max_val] depth. shape (53572,) This shows that the outliers are deleted and the total number of instances in depth is now reduced by the number of outliers. Note that the total number of instances in the Diamond data set is 53,940 and there are 366 outliers. The final number of instances reported is 53,572, which is two more. This is because there are two feature